connorgoggins commented on a change in pull request #17456: Implement remaining
nn_basic ops in opperf
URL: https://github.com/apache/incubator-mxnet/pull/17456#discussion_r372720486
##########
File path: benchmark/opperf/rules/default_params.py
##########
@@ -164,4 +312,5 @@
"mu", "sigma", "lam", "alpha", "beta", "gamma", "k",
"p",
"low", "high", "weight", "bias", "moving_mean",
"moving_var",
"weight", "weight32", "grad", "mean", "var", "mom",
"n", "d",
- "v", "z", "g", "delta", "args"]
+ "v", "z", "g", "delta", "args", "label", "data1",
"data2",
+ "parameters", "state", "loc"]
Review comment:
Yes, those variables do not need to be manually added to the NDArray list.
The way it works is that the underlying arg name ("data") is the only thing
that needs to be in the list, and the mutations of that arg name ("data_*") are
also interpreted as an NDArray.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
